Pith. sign in

REVIEW 3 cited by

When the Curious Abandon Honesty: Federated Learning Is Not Private

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2112.02918 v2 pith:2MHU2GVU submitted 2021-12-06 cs.LG cs.CRcs.DC

classification cs.LGcs.CRcs.DC
keywords datamodelattackweightsgradientsattackercentraldevices
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

In federated learning (FL), data does not leave personal devices when they are jointly training a machine learning model. Instead, these devices share gradients, parameters, or other model updates, with a central party (e.g., a company) coordinating the training. Because data never "leaves" personal devices, FL is often presented as privacy-preserving. Yet, recently it was shown that this protection is but a thin facade, as even a passive, honest-but-curious attacker observing gradients can reconstruct data of individual users contributing to the protocol. In this work, we show a novel data reconstruction attack which allows an active and dishonest central party to efficiently extract user data from the received gradients. While prior work on data reconstruction in FL relies on solving computationally expensive optimization problems or on making easily detectable modifications to the shared model's architecture or parameters, in our attack the central party makes inconspicuous changes to the shared model's weights before sending them out to the users. We call the modified weights of our attack trap weights. Our active attacker is able to recover user data perfectly, i.e., with zero error, even when this data stems from the same class. Recovery comes with near-zero costs: the attack requires no complex optimization objectives. Instead, our attacker exploits inherent data leakage from model gradients and simply amplifies this effect by maliciously altering the weights of the shared model through the trap weights. These specificities enable our attack to scale to fully-connected and convolutional deep neural networks trained with large mini-batches of data. For example, for the high-dimensional vision dataset ImageNet, we perfectly reconstruct more than 50% of the training data points from mini-batches as large as 100 data points.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement

    cs.LG 2026-07 reject novelty 6.0 of 10

    TriShield combines artifact detection, Adam momentum pre-entanglement, and SVD task-subspace projection to drive NeuroImprint reconstruction to 0% with claimed near-zero utility loss.

  2. Gradient Inversion Attack on Graph Neural Networks

    cs.LG 2024-11 conditional novelty 5.0 of 10

    GLG reconstructs node features and graph structure from GNN gradients in federated learning, achieving near-perfect recovery for GraphSAGE and high accuracy for GCN under per-node gradient threat models.

  3. Securing Genomic Data Against Inference Attacks in Federated Learning Environments

    cs.CR 2025-05 conditional novelty 4.0 of 10

    In a synthetic federated learning setup with 100-SNP genomic data, a gradient-norm membership inference attack reached 0.87 F1-score, outperforming confidence-based membership inference and label inference attacks.

Pith tools